Fiber Optic Link

Maybe you can find something here you can use....

https://www.mouser.com/ProductDetail/Analog-Devices/ADL5350-EVALZ?qs=pqQDoYZ2quYEoY9elsf4cw%3D%3D&mgh=1&gclid=Cj0KCQjw_dWGBhDAARIsAMcYuJy7KVlwv6Syi3kv9Mn4CDSJKqUvn6Pd0d6lUT_qgVHPXEdV3zJkmxcaAusvEALw_wcB

An RF Over Fiber (RoF) Link receives an RF Signal, converts it to an optical signal, transmits it over a distance and then converts its back to an RF signal which is then processed as required. A RoF Link consists of the following parts:

• Light Source/Laser Diode - That acts as an optical carrier, usually provided by a laser diode. This laser diode is controlled by forward-biasing the semiconductor junction.
• Electrical-optical Modulator (E/O) - To convert/modulate the light beam using the RF signal.
• Transmission Medium - A single-mode fiber-optic cable. Single-mode has a lower number of light reflections, which lowers attenuation and allows the signal to go further than multimode.
• Optical-electrical Modulator (O/E) - To recover the RF signal at the receiver side, usually by using a photodiode or avalanche photodiode (APD).

https://thorbroadcast.com/product/cellular-rf-signals-over-fiber.html

I looked at the Thor units, but they don't have the frequency range I need.

I'm looking at Mouser, Allied, even Jameco. I can make something, but with the testing and reference baselining would make a project out of that, when I just want to buy something off of the shelf and continue with the current project already in progress.

I have LAN media converters on my desk, and will try to salvage and cobble using that multimode fiber transceiver, but the low end of the frequency range seems to be a limiting factor.

Optical Zonu quoted for me, and would manufacture them for me, but the cost is near $20k...each. I could more easily buy another spec-an with a KVM extender, or a laptop with a dongle and a NIC for less than the cost of one unit!

I thought that this would be existing by 2021 and so common as to be purchased or made for a much lower cost.

Maybe some good used components can be found...and then a frequency lowering mod...

https://www.ebay.com/sch/i.html?_from=R40&_trksid=p2060353.m570.l1313&_nkw=rf+over+fiber+&_sacat=0

Bingo.

What makes you think that converting to optical and back whether using analog or digital techniques to make the transmission, even expensive, will be noiseless and have dynamic range of much real use on an analyser? The most rational way is to use a good low noice and compression RF amplifier at source, to make signal low impedance i,e. 50 Ohm that is more immune ro EMI and convert back to calibrated level just before your instrument. Use your current cable and ground it ONLY on instrument side if (can't understand why would you) run into ground loop problems. S.M.

What is the signal source?

Why not a CAT6 cable?

Cat6 cable will meet your criteria without a fiber link in the middle.

Cat6 must meet very strict noise and cross talk specifications.

If strapping Cat6 to raceways,you must use Velcro ties,not standard cable ties to prevent changing the internal dimensions of the conductors/insulators.

It is a shielded (or non-shielded) twisted pair,or 2 pair. It is relatively cheap,but proper twist must be maintained, even inside the connector itself for max value.

The bandwidth far exceeds your needs,now and in the foreseeable future.

The frequency range is from a minimum of 500 Mhz and can even support 10Gb/sec within a distance of 100 meters(using a 4 conductor channel.

A 100 foot cat6 patch cable,with connectors cost around $18.00.Plain cable is even cheaper.

I wish you good luck.

cuba-pete :

...a low cost option...

might exist , but does not necessarily ensure surprise-/bug -3 operation . . . ? try something Google or Google

I apologize for not visiting the thread back.

To answer a few of your questions and address suggestions, I should have noted that my source cannot be modified else it will require recalibration of a rather sensitive transfer function and likely further reprogramming of a government/contractor processing system. I want to improve my ability to take measurements from my remote office, and not at the equipment itself. Any equipment modding would also put an asterisk on decades-long SNR measurements in myriad publications.

The signal levels I am measuring measure may be in the femtowatts at the source, with a reference of -120 dB in a balanced system. We use a facility reference of -90 dB, where common mode noise already has much higher levels. This is fairly common, so to speak. Interference from an occasional unintended bond or noisy device permits very little common mode noise into my measurements, but it can happen and ruins some. I closely monitor and control the bonding and grounding scheme of installed systems

It’s the radiated noise which is more intrusive over long term measurements and it is that from which I would like to reduce the effect. One of the big offenders is our badge access system which has wiring all over the place. Shielding and choking helps some, but my low levels need better.

Changing one run to my office is much more cost efficient.

Twisted-pair shielded is employed throughout the system I am monitoring, as is balanced coaxial. I don’t have as much control over proximity of the facility wiring as I would like now, and probably less in the future even if I ran conduit to add shielding. If I switched over to fiber, that concern would be gone completely.